Nanoscale sensing elements offer promise for single-molecule analyte detection in physically or biologically constrained environments. Single-walled carbon nanotubes have several advantages when used as optical sensors, such as photostable near-infrared emission for prolonged detection through biological media and single-molecule sensitivity. Molecular adsorption can be transduced into an optical signal by perturbing the electronic structure of the nanotubes. Here, we show that a pair of single-walled nanotubes provides at least four modes that can be modulated to uniquely fingerprint agents by the degree to which they alter either the emission band intensity or wavelength. We validate this identification method in vitro by demonstrating the detection of six genotoxic analytes, including chemotherapeutic drugs and reactive oxygen species, which are spectroscopically differentiated into four distinct classes, and also demonstrate single-molecule sensitivity in detecting hydrogen peroxide. Finally, we detect and identify these analytes in real time within live 3T3 cells, demonstrating multiplexed optical detection from a nanoscale biosensor and the first label-free tool to optically discriminate between genotoxins.

No abstract

Magnetic iron oxide nanoparticles and near-infrared (NIR) fluorescent single-walled carbon nanotubes (SWNT) form heterostructured complexes that can be utilized as multimodal bioimaging agents. Fe catalyst-grown SWNT were individually dispersed in aqueous solution via encapsulation by oligonucleotides with the sequence d(GT) 15 , and enriched using a 0.5 T magnetic array. The resulting nanotube complexes show distinct NIR fluorescence, Raman scattering, and visible/NIR absorbance features, corresponding to the various nanotube species. AFM and cryo-TEM images show DNA-encapsulated complexes composed of a ∼3 nm particle attached to a carbon nanotube on one end. X-ray diffraction (XRD) and superconducting quantum interference device (SQUID) measurements reveal that the nanoparticles are primarily Fe 2 O 3 and superparamagnetic. The Fe 2 O 3 particle-enriched nanotube solution has a magnetic particle content of ∼35 wt %, a magnetization saturation of ∼56 emu/g, and a magnetic relaxation time scale ratio (T 1 /T 2 ) of approximately 12. These complexes have a longer spin−spin relaxation time (T 2 ∼ 164 ms) than typical ferromagnetic particles due to the smaller size of their magnetic component while still retaining SWNT optical signatures. Macrophage cells that engulf the DNA-wrapped complexes were imaged using magnetic resonance imaging (MRI) and NIR mapping, demonstrating that these multifunctional nanostructures could potentially be useful in multimodal biomedical imaging.

Existing temporal pattern mining assumes that events do not have any duration. However, events in many real world applications have durations, and the relationships among these events are often complex. These relationships are modeled using a hierarchical representation that extends Allen's interval algebra. However, this representation is lossy as the exact relationships among the events cannot be fully recovered. In this paper, we augment the hierarchical representation with additional information to achieve a lossless representation. An efficient algorithm called IEMiner is designed to discover frequent temporal patterns from interval-based events. The algorithm employs two optimization techniques to reduce the search space and remove non-promising candidates. From the discovered temporal patterns, we build an interval-based classifier called IEClassifier to differentiate closely related classes. Experiments on both synthetic and real world datasets indicate the efficiency and scalability of the proposed approach, as well as the improved accuracy of IEClassifier.

The cognitive radio (CR) network consists of primary users (PUs) and secondary users (SUs). The SUs in the CR network senses the spectrum band to opportunistically access the white space. Exploiting the white spaces helps to improve the spectrum efficiency. Owing to the excellent learning ability of machine learning/deep learning framework, many works in the recent past have applied shallow/deep multi-layer perceptron approach for spectrum sensing. However, the multi-layer perceptron networks are not well suited for time-series data due to the absence of memory elements. On the other hand, long short-term memory (LSTM) network, an improved version of Recurrent neural network is well suited for time-series data. In this paper, we propose an LSTM based spectrum sensing (LSTM-SS), which learns the implicit features from the spectrum data, for instance, the temporal correlation (i.e., the correlation between the present and past timestamp).Moreover, the CR systems also exploits the PU activity statistics to improve the CR performance. In this context, we compute the PU activity statistics like on and off period duration, duty cycle and propose the PU activity statistics based spectrum sensing (PAS-SS) to enhance the sensing performance. The proposed sensing schemes are validated on the spectrum data of various radio technologies acquired using an experimental test-bed setup. The proposed LSTM-SS scheme is compared with the state of the art spectrum sensing techniques. Experimental results indicate that the proposed schemes has improved detection performance and classification accuracy at low signal to noise ratio regimes. We notice that the improvement achieved is at the cost of longer training time and a nominal increase in execution time.

No abstract

No abstract

Copy number alterations (CNAs) are thought to account for 85% of the variation in gene expression observed among breast tumours. The expression of cis-associated genes is impacted by CNAs occurring at proximal loci of these genes, whereas the expression of trans-associated genes is impacted by CNAs occurring at distal loci. While a majority of these CNA-driven genes responsible for breast tumourigenesis are cis-associated, trans-associated genes are thought to further abet the development of cancer and influence disease outcomes in patients. Here we present a network-based approach that integrates copy-number and expression profiles to identify putative cis-and trans-associated genes in breast cancer pathogenesis. We validate these cis-and trans-associated genes by employing them to subtype a large cohort of breast tumours obtained from the METABRIC consortium, and demonstrate that these genes accurately reconstruct the ten subtypes of breast cancer. We observe that individual breast cancer subtypes are driven by distinct sets of cis-and trans-associated genes. Among the cisassociated genes, we recover several known drivers of breast cancer (e.g. CCND1, ERRB2, MDM2 and ZNF703) and some novel putative drivers (e.g. BRF2 and SF3B3). siRNAmediated knockdown of BRF2 across a panel of breast cancer cell lines showed significant reduction in cell viability for ER-/HER2+ (MDA-MB-453) cells, but not in normal (MCF10A) cells thereby indicating that BRF2 could be a viable therapeutic target for ER-/HER2+ cancers. Among the trans-associated genes, we identify modules of immune-response (CD2, CD19, CD38 and CD79B), mitotic/cell-cycle kinases (e.g. AURKB, MELK, PLK1 and TTK), and DNA-damage response genes (e.g. RFC4 and FEN1). siRNA-mediated knockdown of RFC4 significantly reduced cell proliferation in estrogen receptor-negative normal breast and cancer lines, thereby indicating that RFC4 is essential for cancer cell survival but could also be a useful biomarker for aggressive (ER-negative) breast tumours. Availability: http://bioinformatics.org.au/tools-data/ under NetStrat.